First "Cold Fusion" article, 1926!

[Dinu Popa]

Review of article by F. A. Paneth and K. Peters; published in Nature, May 14 1927

The current (September) issue of "Berichte" of the German Chemical Society contains a paper by Profs. F. Paneth and K. Peters on "The Transformation of Hydrogen into Helium," in which they describe in outline how they have succeeded in detecting the presence of very minute amounts of helium, of the order of one hundred millionth of a cubic centimeter, derived from hydrogen which had been absorbed by finely divided palladium at the ordinary temperature.

Theory indicating that this conversion would involve the liberation of much energy (6.4 x 10¹¹ cal. from 4 gram-atoms of hydrogen), the author's primary task was to find out if the change would take place without introducing energy from outside, e.g. in the presence of a catalyst; and in order to be able to detect very small quantities of helium they elaborated the spectroscopic method in such a way that the limiting amount detectable was 10^-8 - 10^-9 cc, or 10^-12 - 10^-13 gm. Easily liquefiable gas was removed with liquid air and charcoal; oxygen was added and the hydrogen burnt on the surface of the catalyst; water-vapour and excess oxygen were removed with charcoal, and the residual gas was passed into a glass capillary-tube of 0.1 mm section, which was surrounded with electro-wires and placed before the slit of the spectroscope. Every precaution was taken to exclude atmosphere helium; the portion of the apparatus that was heated was surrounded with a vacuum-mantle and immersed in water. The presence of neon lines afforded a most valuable criterion of the presence of atmospheric gases; neon was never completely excluded, but the amount present was so small that it did not invalidate the author's main conclusion. The method is so delicate that the liberation of helium from a mixture of thorium B and thorium C was easily detected, while it is sufficiently sensitive to determine the presence of helium in a few cubic centimeters of natural gas. By its means a natural gas containing 0.19 percent by volume of helium was discovered in Germany, and steps have been taken to exploit it commercially. The Canadian natural gas from which helium is extracted contains 0.33 percent by volume.

Attempts were made to effect the transformation by submitting hydrogen to the action of a silent electro-discharge in an ozone apparatus, and by passing a prolonged and powerful discharge through it in a Geissler-tube fitted with aluminum electrodes; but no success was achieved. Nor was the attempt to produce helium by bombarding certain salts with cathode rays, as suggested by Lord Rayleigh, any more fertile, so that recourse was had to passing fairly large amounts of hydrogen - up to one liter - through heated palladium, in the hope that at the moment of exit a fraction of the protons and electrons would combine to form the helium nucleus. In this case the indications were favorable, but the result was inconclusive owing to the presence of atmospheric neon, and the absence of any proportionality between the strength of the helium lines and the amount of hydrogen that was used.

Finely divided palladium, either as sponge, 'black', or palladinised asbestos, was then used to absorb hydrogen at room temperature, and after different intervals of time the hydrogen was combined with oxygen, as previously described. The residual gas obtained after a 12-hours' contact between palladium and hydrogen exhibited four or five lines of the helium spectrum and no neon lines; there was also a distinct proportionality between the amount of helium observed and the duration of the time of contact. The activity of the different palladium preparations employed varied considerably; it invariably diminished with repeated use, but both the power of absorbing hydrogen and of effecting the transformation were restored by heating the hydrogen or oxygen, in a mixture of these gases, or in a vacuum. No helium production was observed with palladium preparations that did not absorb hydrogen, although preparations were occasionally obtained that absorbed hydrogen well but gave little or no helium, especially if the hydrogen had been absorbed at a high temperature.

The above results indicated that palladium preparations that have long remained unused at room temperature should contain a little helium (not of atmospheric origin). Examination of a number of such specimens showed that helium was present in all of them, and in particular a specimen of palladinised asbestos, which had been purchased from Kahlbaum two years previously, was found to be relatively very rich, 1 gm of it containing 10^-6 cc of practically pure helium. After this specimen had been heated to expel the helium, and treated with oxygen for twelve hours, no fresh helium was detected, but at the end of five hours in contact with hydrogen a considerable amount of helium was found. This experiment was performed three times with the same result. The palladium, however, gradually lost its activity; at the beginning it produced helium at the rate of 10^-8 - 10^-7 cc per day; after twenty treatments it became inactive. Its activity was restored, although not to the original degree, in the manner described above. Finely divided platinum is less active than palladium, and the action of pyrophoric nickel is weaker still.

The authors discuss fully the possible sources of error in their experiments, such as the ingress of atmospheric helium, the adsorptive capacity of glass for helium, the conceivable preferential adsorption of helium by palladium, or by asbestos, and the possibility of helium being formed as a radio-active disintegration product of palladium; all of which they consider to be excluded. The hydrogen and oxygen used by them contained less than 0.001 per cent of air. They were not able to detect any trace of the energy liberated during the transformation, and they point out that the amount set free from the conversion of such small quantities of hydrogen - about 0.28 calorie - would be extremely difficult to detect, and particularly so if thermal changes due to absorption or formation of compounds also take place. They incline to the view that the liberated energy is more likely to appear as radiation, e.g. as gamma or Millikan-rays (cosmic rays), than as heat.

(These authors later acknowledged that they could not measure He with enough precision in order to be able to draw a conclusion)

[Dinu Popa]

A few months ago, K. Peters and I published an account of experiments we had made in an attempt to transmute hydrogen into helium ("Ber. d. Deutschen Chem. Ges.",(vol) 59, 2039, 1926). A more or less detailed account of this publication appeared in the columns of Nature (vol. 118, p. 526, 1926), and perhaps I may be permitted to refer to a more recent publication on the same topic by K. Peters, P. Gunther, and myself ("Ber. d. Deutschen Chem. Ges.", (vol) 60, 808, 1927). In this communication, as a result of further experiments, we feel that we are in a position to give an explanation of the occurrence of the observed very small quantities of helium in our experiments, without having recourse to the assumption of a synthesis of helium.

In the first-mentioned communication we considered the penetration of helium from the atmosphere through the glass walls of the apparatus to be the most likely source of trouble in such experiments, and we excluded this possibility by the use of vacuum jackets, immersion in water, and similar devices. In addition, we also discussed the possibility of regarding the helium dissolved in the glass as an explanation of the observed effects, but blank experiments led us to the conclusion that the quantity of helium capable of being liberated in this way was beyond the limits of sensitivity of our method of detection. In the interval we have carried out experiments both in the Baker Laboratory of Cornell and in the Chemical Laboratory of the University of Berlin, and these have shown that the liberation of helium from glass (and from asbestos) is dependent on the presence of hydrogen. Thus glass tubes which gave off no detectable quantities of helium when they were heated in a vacuum or in oxygen were found to yield helium in quantities of the order of 10-9 cc when they were heated in an atmosphere of hydrogen. Now in the earlier experiments the glass tubes containing palladium yielded helium, whereas the empty glass tubes used in control experiments did not; and since the former tubes would fill hydrogen on the application of heat, we see that the source of helium lay not in the palladium but in the glass, in spite of appearances to the contrary.

Our method of detecting helium is sufficiently sensitive to show that a glass tube which has been completely freed from its content of helium by heating in hydrogen takes up a detectable amount of neon-free helium from the atmosphere even after only one day's contact with the air.

Since asbestos behaves similarly to glass, we now see why one particular palladium preparation, bought as palladium-asbestos, yielded large quantities (10^-7 cc) of helium after being charges with hydrogen. Here, obviously, in contrast to the preparations we made ourselves, the asbestos had not been ignited until it was free from helium, and a fraction of the residual helium was always liberated by heating when the palladium was charged with hydrogen, whereas in oxygen no development of helium could be observed.

As a result of our more recent experiments we have thus established that, in using an apparatus made of glass, one cannot make any trustworthy statement as to the origin of 10^-9 cc of helium if air comes in contact with the apparatus, parts of which are later heated in hydrogen. By avoiding all heating of the apparatus, we shall endeavor to decide whether a transmutation of hydrogen into helium of the order of 10^-9 cc or less takes place. In any case, the amount of helium formed in experiments on electric discharges, as tested by various workers and by ourselves, and in experiments on the action of palladium, does not reach the order of magnitude of 10^-8 cc.

It is scarcely necessary to emphasize the fact that the sensitiveness of our method, though limited to 10^-8 cc, is sufficient to decide with certainty the other questions dealt with in our first communication, such as the helium content of meteorites, the helium development of radioactive deposits, and so on.

Fritz Paneth, Berlin, Mar. 2